Chronic fatigue syndrome (CFS) has sparked growing interest in recent decades, despite which the condition continues to be controversial. The main reason for this is the lack of objective evidence of the syndrome. CFS is a heterogeneous multisystem disorder of unknown origin characterised by generalised fatigue lasting over 6 months, worsening after exercise (98%), and associated with recurrent headache (90%) and concentration and/or memory problems (85%).1 These symptoms are usually associated with non-restorative sleep (94%), musculoskeletal pain (75%), swollen lymph nodes, and/or psychiatric disorders (65%).1 Prevalence of CFS is estimated at 0.007%-2.8% of the general adult population, predominantly affecting women (female-to-male ratio of 4:1).2

The current concept of CFS overlaps with such other syndromes as fibromyalgia, temporomandibular pain, irritable bowel syndrome, and sleep disorders (SD). According to the third edition of the International Classification of Sleep Disorders (ICSD-3),3 over 90 well-characterised SDs have been described. Some of these disorders are highly prevalent among the general population: insufficient sleep syndrome (40%), insomnia (20%), sleep-related breathing disorders (2%-5%), restless legs syndrome (RLS) (2%), periodic limb movement disorder (PLMD), delayed sleep-wake phase disorder, and parasomnia.3 SDs have also been associated with increased mortality; a higher prevalence of cardiovascular disease, workplace and traffic accidents, and psychiatric disorders; and poorer cognitive function and performance.4,5

SDs and CFS share several symptoms, such as fatigue and non-restorative sleep, both of which are subjective and currently impossible to quantify. Non-restorative sleep, one of the most common symptoms in patients with CFS (87%-95%),6,7 constitutes a key symptom for CFS diagnosis.8 The latest diagnostic guidelines for CFS include such other symptoms as excessive daytime sleepiness (EDS), difficulty falling asleep, and sleep fragmentation7; these symptoms are also very common in patients with SDs.3 The clinical overlap between SDs and CFS, combined with the current lack of knowledge of SDs, may explain the delayed diagnosis in these entities.

A comprehensive review of the topic was conducted in order to address the following questions: (1) what symptoms of SDs overlap with those of CFS? (2) what is the prevalence of SD in patients with CFS? (3) what polysomnographic alterations have been observed in CFS? and (4) how can SDs be diagnosed in patients with CFS and differentiated from the latter condition? To aid in clinical differentiation between these disorders, useful clinical guidelines for SD screening are provided, together with a diagnostic algorithm for SDs in patients with CFS, according to the sleep pattern they present. Identifying SDs in patients with suspected CFS will provide new diagnostic and treatment strategies to improve these patients’ quality of life.

A systematic literature search was conducted on MEDLINE (PubMed) using the MeSH terms [sleep]+[chronic fatigue syndrome]. Inclusion criteria were as follows: articles published in indexed journals, written in Spanish or English, and published prior to 31 December 2014. The study includes original articles reporting prospective and retrospective studies, systematic reviews, and meta-analyses. After reading the abstracts and selecting those articles that appeared to be relevant, the full texts of the selected articles, and the cited references, were reviewed.

Fatigue, EDS, and non-restorative sleep are very common symptoms of CFS.8 However, the diagnostic criteria for many SDs defined in the ICSD-33 (insomnia, sleep-related breathing disorders, central disorders of hypersomnolence, circadian rhythm sleep-wake disorders, and sleep-related movement disorders) greatly coincide with those of CFS, which makes it more difficult to differentiate the 2 conditions.

Since 1988, diagnostic criteria for CFS have always included very frequent symptoms of primary SDs (Table 1). The 1988 (Holmes et al.9) and 1994 (Fukuda et al.10) Centers for Disease Control and Prevention (CDC) criteria include non-restorative sleep as a minor symptom. Furthermore, the diagnostic criteria proposed by Fukuda et al. include obstructive sleep apnoea/hypopnoea syndrome (OSAHS) and narcolepsy as exclusion criteria for CFS. The Canadian criteria, published in 2003, list non-restorative sleep as a major diagnostic criterion, and also include such other symptoms as EDS, alterations in sleep duration, and sleep-wake cycle alterations.1 These criteria also list OSAHS, central sleep apnoea, RLS, and PLMD as exclusion criteria. Lastly, the 2011 International Consensus Criteria1 list sleep-related symptoms within the neurological alterations section; these include insomnia, EDS, non-restorative sleep, long naps, sleep fragmentation, nightmares, and vivid dreams. Although the 2011 International Consensus did not include any SD among the exclusion criteria, it emphasised that these disorders may be present and should be considered as a comorbidity of CFS, underscoring the importance of identifying and treating each entity independently.

As mentioned previously, many of the diagnostic criteria for CFS include such symptoms as fatigue, EDS, and non-restorative sleep; however, these terms are often ill-defined or confusing. Therefore, the most standardised definition of each term is provided, with a view to using these terms correctly and uniformly.

According to the published evidence, the mean prevalence of sleep disorders in patients with CFS is estimated at 56% (range 18%-81%) (Table 2).8,19,21–24,31–36 Some authors also describe a greater level of functional impairment in patients with CFS plus an associated primary SD.32

These studies reporting a higher prevalence of SD, however, have a number of limitations: they all include a small number of patients, and tend only to describe subjective clinical manifestations and to rule out only primary SD (OSAHS and/or narcolepsy and/or PLMD); and the patients included were referred to a sleep unit for assessment. These limitations may have affected the accuracy of the findings regarding SD prevalence.

These studies mainly follow the diagnostic criteria established by Fukuda et al.,10 according to which OSAHS and narcolepsy must be excluded for CFS to be diagnosed; despite this, however, these 2 disorders are highly prevalent among the populations included in these studies. Other relatively frequent primary SDs include chronic insomnia, primary hypersomnia, insufficient sleep syndrome, poor sleep hygiene, and circadian rhythm sleep-wake disorders. However, some studies have reported pathological levels of EDS (<8min in MSLT findings) in 22%-41% of patients,23,24,31,33 which suggests that these patients have such conditions as primary hypersomnia, insufficient sleep syndrome, or poor sleep hygiene. The same is true for chronic insomnia, with an estimated prevalence of 68%-86%, according to some studies.19,35

This explains the current debate around the association between SDs and CFS. According to some researchers, presence of SDs should rule out CFS; others, however, regard SDs and CFS as concomitant disorders. There is insufficient evidence to confirm the hypothesis that SDs are concomitant to CFS. Firstly, Nisenbaum et al.8 conducted a population study into CFS over the course of 3 years. According to their results, diagnostic criteria for CFS persisted throughout the study period in 20%-33% of the patients and fully or partially remitted in 56%; 23% received alternative diagnoses (SDs in 20%: OSAHS and narcolepsy). Secondly, the concomitant disease hypothesis was introduced by Libman et al.34 in a study including patients with CFS and associated OSAHS. These authors compared patients displaying good compliance with CPAP to non-compliant patients, finding no significant differences in CFS symptoms. This study included a very small sample and did not provide an objective definition of treatment compliance. However, it did find non-significant improvements; more promising results may have been achieved if treatment compliance had been monitored more thoroughly and the sample had been larger. In another study describing 4 patients with OSAHS and CFS who received treatment with CPAP, compliant patients displayed total symptom resolution, whereas non-compliant and partially compliant patients displayed no improvements.37 Despite the small sample size, the results of the latter study are more in line with those published by Tomfohr et al.,38 who conducted a randomised placebo-controlled study of patients diagnosed with OSAHS, finding that patients treated with CPAP manifested significant improvements in fatigue compared to those receiving placebo CPAP.

According to the Canadian diagnostic criteria for CFS, identifying SDs is essential in patients with suspected CFS, since the presence of these disorders should be regarded as an exclusion criterion for CFS. It is therefore logical to consider that the different SDs associated with fatigue and CFS may improve significantly with adequate treatment. Only in the event that SDs do not improve and/or residual symptoms meeting the criteria for CFS persist should patients be re-evaluated and SDs be considered as a comorbidity of CFS.

The above suggests high prevalence of primary SDs among patients with CFS and that symptoms of CFS overlap to a great extent with those of SDs. Patients with CFS normally have poor sleep quality, reported as a feeling of non-restorative sleep. This is a rather ambiguous concept; presence of non-restorative sleep should not be described in binary terms of presence/absence. Physicians should aim to gather as much information as possible in order to better define and manage the underlying SDs. In view of this, physicians should ask patients whether they would describe their sleep as non-restorative and enquire into their sleep-wake schedule, number of hours of sleep, number of intra-sleep awakenings, intra-sleep wakefulness time, and subjective sleep efficiency. They should also enquire about the impact of non-restorative sleep on daytime activity, gathering data on fatigue, feelings of excessive daytime sleepiness, the need to take naps, and the number and duration of these naps. Other relevant data include snoring, sleep apnoea, movements during sleep, nightmares, nocturnal sweating, nocturia, and/or nocturnal reflux (Table 4).

Primary SDs have been reported to be highly prevalent among patients with CFS. Sleep-related breathing disorders are highly prevalent in the general population (2%-5%)4; this rules out OSAHS, for which a highly effective treatment is available, in all patients with CFS. These patients may also present other less well-known but also highly prevalent primary SDs whose specific treatment may improve patients’ quality of life; these include chronic insomnia, PLMD, RLS, circadian rhythm sleep-wake disorders (delayed or advanced sleep-wake phase disorder and/or irregular sleep-wake rhythm disorder), parasomnias (sleepwalking, REM sleep behaviour disorder, and nightmares), poor sleep hygiene, insufficient sleep syndrome, narcolepsy, and/or idiopathic hypersomnia. Gathering basic data on patients’ sleeping habits and history of SDs is therefore essential in screening for any underlying SD (Table 4).

Gotts et al.36 described 4 sleep phenotypes in patients with CFS: deep sleepers, light sleepers, light sleepers with daytime activities affected, and insomniacs (conciliation, maintenance, and/or early awakening). Based on these data, a diagnostic algorithm is proposed for the differential diagnosis of SDs in patients with CFS (Fig. 1).

Figure 1.

Algorithm for the differential diagnosis of different sleep disorders associated with chronic fatigue syndrome. PLMD: periodic limb movement disorder; CSAS: central sleep apnoea syndrome; OSAHS: obstructive sleep apnoea/hypopnoea syndrome.

(0.34MB).

Symptoms of CFS overlap with those of many other disorders, particularly SDs; this explains why CFS is frequently misdiagnosed as other conditions. Multiple researchers now apply the Fukuda et al. criteria, despite which a high prevalence of primary SDs has been reported. Therefore, with a view to enabling differentiation between CFS and SDs, greater effort should be made to properly define such symptoms as fatigue, EDS, and non-restorative sleep in order to avoid confusion and overlap between symptoms. The concept of non-restorative sleep, which is widely used in CFS, should only be applied to patients who do not feel refreshed upon awakening despite sleeping sufficient hours, after primary SDs have been ruled out. SDs frequently associated with CFS (OSAHS and narcolepsy) must be ruled out in these patients. The presence of other highly prevalent SDs should also be assessed; these include insomnia, insufficient sleep syndrome, poor sleep hygiene, primary hypersomnia, circadian rhythm sleep disorders, PLMD, rhythmic movement disorder, and parasomnias, all of which are very frequently associated with CFS (Fig. 1). Several tools are currently available to diagnose these disorders, including polysomnography and the MSLT; likewise, a number of effective treatments have been developed to improve patients’ quality of life and reduce morbidity and mortality.

Some researchers have suggested that SDs are comorbidities of CFS; it is this author's view, however, that SDs should be regarded as primary disorders, distinct from CFS, and considered an exclusion criterion for CFS diagnosis. Nonetheless, an SD may be considered a comorbidity of CFS when diagnostic criteria for CFS persist despite treatment in a patient with a diagnosis of a primary SD; a re-evaluation is necessary in these cases.

Studies of sleep in patients with CFS are scarce, heterogeneous, and provide conflicting results. Future research should aim to improve patient selection by better identifying concomitant primary SDs and providing a more detailed, uniform description of symptoms. This would produce more homogeneous, conclusive data.

None declared.

The author has no conflicts of interest to declare. This study has received no funding.